metadata
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
model-index:
- name: emotion_classification
results:
- task:
name: Image Classification
type: image-classification
dataset:
name: imagefolder
type: imagefolder
config: default
split: train
args: default
metrics:
- name: Accuracy
type: accuracy
value: 0.54375
emotion_classification
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.3248
- Accuracy: 0.5437
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine_with_restarts
- lr_scheduler_warmup_steps: 18
- num_epochs: 18
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 1.0 | 20 | 1.4238 | 0.4625 |
| No log | 2.0 | 40 | 1.4238 | 0.4188 |
| No log | 3.0 | 60 | 1.4457 | 0.4625 |
| No log | 4.0 | 80 | 1.5008 | 0.45 |
| No log | 5.0 | 100 | 1.4511 | 0.5062 |
| No log | 6.0 | 120 | 1.6197 | 0.4375 |
| No log | 7.0 | 140 | 1.6023 | 0.45 |
| No log | 8.0 | 160 | 1.6301 | 0.4562 |
| No log | 9.0 | 180 | 1.7171 | 0.4688 |
| No log | 10.0 | 200 | 1.9459 | 0.3688 |
| No log | 11.0 | 220 | 1.8110 | 0.4062 |
| No log | 12.0 | 240 | 1.7246 | 0.425 |
| No log | 13.0 | 260 | 1.7258 | 0.475 |
| No log | 14.0 | 280 | 1.7419 | 0.475 |
| No log | 15.0 | 300 | 1.6547 | 0.4625 |
| No log | 16.0 | 320 | 1.7231 | 0.4625 |
| No log | 17.0 | 340 | 1.8635 | 0.4125 |
| No log | 18.0 | 360 | 1.6725 | 0.4562 |
Framework versions
- Transformers 4.41.1
- Pytorch 2.1.2
- Datasets 2.19.1
- Tokenizers 0.19.1